# The Islamia University of Bahawalpur # Submitted By: MUHAMMAD SHARIQ # Submitted By: MUHAMMAD ATEEQ # Reg No: F21BDATS1M02050 # BS DATA SCIENCE 4rth Semester Morning-B # Machine Learning Project # Heart Failure Prediction Dataset

import numpy as np
import pandas as pd
import tensorflow as tf
import matplotlib as plt
import seaborn as sns
from tensorflow.keras.models import Sequential 
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.layers import Dense
from sklearn.model_selection import train_test_split

Loading the Dataset

heart_data = pd.read_csv(r"heart.csv")
heart_data.sample(10)

	Age	Sex	ChestPainType	RestingBP	Cholesterol	FastingBS	RestingECG	MaxHR	ExerciseAngina	Oldpeak	ST_Slope	HeartDisease
604	68	M	NAP	134	254	1	Normal	151	Y	0.0	Up	0
374	61	M	ASY	125	0	0	Normal	105	Y	0.0	Down	1
11	58	M	ATA	136	164	0	ST	99	Y	2.0	Flat	1
799	53	M	NAP	130	246	1	LVH	173	N	0.0	Up	0
522	50	M	ASY	144	349	0	LVH	120	Y	1.0	Up	1
848	52	M	ASY	128	255	0	Normal	161	Y	0.0	Up	1
199	57	F	TA	130	308	0	Normal	98	N	1.0	Flat	0
291	47	F	ATA	140	257	0	Normal	135	N	1.0	Up	0
702	59	M	TA	178	270	0	LVH	145	N	4.2	Down	0
771	55	M	ASY	140	217	0	Normal	111	Y	5.6	Down	1

heart_data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 918 entries, 0 to 917
Data columns (total 12 columns):
 #   Column          Non-Null Count  Dtype  
---  ------          --------------  -----  
 0   Age             918 non-null    int64  
 1   Sex             918 non-null    object 
 2   ChestPainType   918 non-null    object 
 3   RestingBP       918 non-null    int64  
 4   Cholesterol     918 non-null    int64  
 5   FastingBS       918 non-null    int64  
 6   RestingECG      918 non-null    object 
 7   MaxHR           918 non-null    int64  
 8   ExerciseAngina  918 non-null    object 
 9   Oldpeak         918 non-null    float64
 10  ST_Slope        918 non-null    object 
 11  HeartDisease    918 non-null    int64  
dtypes: float64(1), int64(6), object(5)
memory usage: 86.2+ KB

heart_data.isnull().sum()

Age               0
Sex               0
ChestPainType     0
RestingBP         0
Cholesterol       0
FastingBS         0
RestingECG        0
MaxHR             0
ExerciseAngina    0
Oldpeak           0
ST_Slope          0
HeartDisease      0
dtype: int64

Data Preprocessing

heart_data.columns

Index(['Age', 'Sex', 'ChestPainType', 'RestingBP', 'Cholesterol', 'FastingBS',
       'RestingECG', 'MaxHR', 'ExerciseAngina', 'Oldpeak', 'ST_Slope',
       'HeartDisease'],
      dtype='object')

heart_data.describe()

	Age	RestingBP	Cholesterol	FastingBS	MaxHR	Oldpeak	HeartDisease
count	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000
mean	53.510893	132.396514	198.799564	0.233115	136.809368	0.887364	0.553377
std	9.432617	18.514154	109.384145	0.423046	25.460334	1.066570	0.497414
min	28.000000	0.000000	0.000000	0.000000	60.000000	-2.600000	0.000000
25%	47.000000	120.000000	173.250000	0.000000	120.000000	0.000000	0.000000
50%	54.000000	130.000000	223.000000	0.000000	138.000000	0.600000	1.000000
75%	60.000000	140.000000	267.000000	0.000000	156.000000	1.500000	1.000000
max	77.000000	200.000000	603.000000	1.000000	202.000000	6.200000	1.000000

heart_data["Sex"] = heart_data["Sex"].replace({"F":0,'M':1})
heart_data["ChestPainType"] = heart_data["ChestPainType"].replace({"ASY":0,"ATA":1,"NAP":2,"TA":3})
heart_data["RestingBP"] = np.where(heart_data["RestingBP"] < 120, 0, 1)
heart_data["ExerciseAngina"] = heart_data["ExerciseAngina"].replace({"N":0,'Y':1})
heart_data["RestingECG"] = heart_data["RestingECG"].replace({'Normal':0,"ST":1,"LVH":2})
heart_data["ST_Slope"] = heart_data["ST_Slope"].replace({"Flat":0,'Up':1})
heart_data.head(10)

	Age	Sex	ChestPainType	RestingBP	Cholesterol	FastingBS	RestingECG	MaxHR	ExerciseAngina	Oldpeak	ST_Slope	HeartDisease
0	40	1	1	1	289	0	0	172	0	0.0	1	0
1	49	0	2	1	180	0	0	156	0	1.0	0	1
2	37	1	1	1	283	0	1	98	0	0.0	1	0
3	48	0	0	1	214	0	0	108	1	1.5	0	1
4	54	1	2	1	195	0	0	122	0	0.0	1	0
5	39	1	2	1	339	0	0	170	0	0.0	1	0
6	45	0	1	1	237	0	0	170	0	0.0	1	0
7	54	1	1	0	208	0	0	142	0	0.0	1	0
8	37	1	0	1	207	0	0	130	1	1.5	0	1
9	48	0	1	1	284	0	0	120	0	0.0	1	0

heart_data.sample(10)

	Age	Sex	ChestPainType	RestingBP	Cholesterol	FastingBS	RestingECG	MaxHR	ExerciseAngina	Oldpeak	ST_Slope	HeartDisease
481	69	1	2	1	0	1	1	118	0	2.5	Down	1
887	43	1	0	1	247	1	2	143	1	0.1	0	1
130	38	1	2	1	292	0	0	130	0	0.0	1	0
57	58	1	2	1	213	0	1	140	0	0.0	0	1
874	43	1	0	1	247	0	0	171	0	1.5	1	0
513	62	1	3	0	258	0	1	150	1	1.3	0	1
145	39	1	0	0	273	0	0	132	0	0.0	1	0
263	59	1	0	1	126	0	0	125	0	0.0	0	1
131	46	1	0	0	202	0	0	150	1	0.0	0	1
655	40	1	0	1	223	0	0	181	0	0.0	1	1

Data Analyzing

import matplotlib.pyplot as plt
categorical_vars = ['Sex', 'ChestPainType', 'FastingBS', 'RestingECG', 'ExerciseAngina', 'ST_Slope']

plt.figure(figsize=(12, 8))
for i, var in enumerate(categorical_vars):
    plt.subplot(2, 3, i+1)
    sns.countplot(data=heart_data, x=var)
    plt.xlabel(var)
plt.tight_layout()
plt.show()

sns.countplot(data=heart_data, x="HeartDisease",hue="Sex")

<Axes: xlabel='HeartDisease', ylabel='count'>

heart_data_corr = heart_data[["HeartDisease","Sex","ChestPainType","RestingBP", "FastingBS", "RestingECG", "ExerciseAngina"]].corr()
heart_data_corr
plt.figure(figsize = (20,10))
sns.heatmap(heart_data_corr, cmap='crest', annot=True)

<Axes: >

import matplotlib.pyplot as plt
plt.figure(figsize=(13, 7))  
sns.boxplot(data=heart_data_corr)

<Axes: >

heart_data.describe()

	Age	Sex	ChestPainType	RestingBP	Cholesterol	FastingBS	RestingECG	MaxHR	ExerciseAngina	Oldpeak	HeartDisease
count	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000
mean	53.510893	0.789760	0.781046	0.824619	198.799564	0.233115	0.603486	136.809368	0.404139	0.887364	0.553377
std	9.432617	0.407701	0.956519	0.380500	109.384145	0.423046	0.805968	25.460334	0.490992	1.066570	0.497414
min	28.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	60.000000	0.000000	-2.600000	0.000000
25%	47.000000	1.000000	0.000000	1.000000	173.250000	0.000000	0.000000	120.000000	0.000000	0.000000	0.000000
50%	54.000000	1.000000	0.000000	1.000000	223.000000	0.000000	0.000000	138.000000	0.000000	0.600000	1.000000
75%	60.000000	1.000000	2.000000	1.000000	267.000000	0.000000	1.000000	156.000000	1.000000	1.500000	1.000000
max	77.000000	1.000000	3.000000	1.000000	603.000000	1.000000	2.000000	202.000000	1.000000	6.200000	1.000000

from pandas_profiling import ProfileReport
heart_data = pd.read_csv(r"heart.csv")
report = ProfileReport(heart_data)
display(report)

C:\Users\AFC\AppData\Local\Temp\ipykernel_1460\363051215.py:1: DeprecationWarning: `import pandas_profiling` is going to be deprecated by April 1st. Please use `import ydata_profiling` instead.
  from pandas_profiling import ProfileReport

Summarize dataset:   0%|          | 0/5 [00:00<?, ?it/s]

Generate report structure:   0%|          | 0/1 [00:00<?, ?it/s]

Render HTML:   0%|          | 0/1 [00:00<?, ?it/s]

Featuring and Labeling Data

x = heart_data[["Sex","ChestPainType","RestingBP", "FastingBS", "RestingECG", "ExerciseAngina"]]
y = heart_data["HeartDisease"]

Spliting the Data

X_train, X_test, Y_train, Y_test = train_test_split(x,y, test_size = 0.9, random_state = 42)

Modeling the Data

model = Sequential([Dense(64, activation = "relu"),
                   Dense(32, activation = "relu"),
                   Dense(1, activation = "sigmoid")
                   ])

Choosing the Optimizer, Loss Function and Fitting the Model

optimizer = Adam(learning_rate=0.01)
model.compile(loss = "binary_crossentropy", optimizer = optimizer, metrics = ["accuracy"])
history = model.fit(X_train, Y_train, validation_data=(X_test, Y_test), epochs=100)

Epoch 1/100
3/3 [==============================] - 1s 111ms/step - loss: 0.6864 - accuracy: 0.5165 - val_loss: 0.6179 - val_accuracy: 0.7400
Epoch 2/100
3/3 [==============================] - 0s 32ms/step - loss: 0.6144 - accuracy: 0.7363 - val_loss: 0.5779 - val_accuracy: 0.7703
Epoch 3/100
3/3 [==============================] - 0s 33ms/step - loss: 0.5701 - accuracy: 0.7253 - val_loss: 0.5241 - val_accuracy: 0.7763
Epoch 4/100
3/3 [==============================] - 0s 34ms/step - loss: 0.5360 - accuracy: 0.7473 - val_loss: 0.5342 - val_accuracy: 0.7775
Epoch 5/100
3/3 [==============================] - 0s 30ms/step - loss: 0.5152 - accuracy: 0.7582 - val_loss: 0.5395 - val_accuracy: 0.7352
Epoch 6/100
3/3 [==============================] - 0s 31ms/step - loss: 0.4850 - accuracy: 0.7582 - val_loss: 0.5277 - val_accuracy: 0.7823
Epoch 7/100
3/3 [==============================] - 0s 35ms/step - loss: 0.4806 - accuracy: 0.7582 - val_loss: 0.5685 - val_accuracy: 0.7485
Epoch 8/100
3/3 [==============================] - 0s 28ms/step - loss: 0.4584 - accuracy: 0.7912 - val_loss: 0.5674 - val_accuracy: 0.7437
Epoch 9/100
3/3 [==============================] - 0s 33ms/step - loss: 0.4500 - accuracy: 0.7912 - val_loss: 0.5846 - val_accuracy: 0.7400
Epoch 10/100
3/3 [==============================] - 0s 35ms/step - loss: 0.4442 - accuracy: 0.8022 - val_loss: 0.6168 - val_accuracy: 0.7437
Epoch 11/100
3/3 [==============================] - 0s 34ms/step - loss: 0.4353 - accuracy: 0.8022 - val_loss: 0.6072 - val_accuracy: 0.7388
Epoch 12/100
3/3 [==============================] - 0s 30ms/step - loss: 0.4262 - accuracy: 0.8022 - val_loss: 0.6172 - val_accuracy: 0.7437
Epoch 13/100
3/3 [==============================] - 0s 40ms/step - loss: 0.4250 - accuracy: 0.8022 - val_loss: 0.6298 - val_accuracy: 0.7437
Epoch 14/100
3/3 [==============================] - 0s 40ms/step - loss: 0.4128 - accuracy: 0.8132 - val_loss: 0.6422 - val_accuracy: 0.7461
Epoch 15/100
3/3 [==============================] - 0s 40ms/step - loss: 0.4014 - accuracy: 0.8132 - val_loss: 0.6283 - val_accuracy: 0.7461
Epoch 16/100
3/3 [==============================] - 0s 39ms/step - loss: 0.3954 - accuracy: 0.8132 - val_loss: 0.6252 - val_accuracy: 0.7437
Epoch 17/100
3/3 [==============================] - 0s 47ms/step - loss: 0.3920 - accuracy: 0.8242 - val_loss: 0.6584 - val_accuracy: 0.7437
Epoch 18/100
3/3 [==============================] - 0s 36ms/step - loss: 0.3815 - accuracy: 0.8242 - val_loss: 0.6702 - val_accuracy: 0.7412
Epoch 19/100
3/3 [==============================] - 0s 27ms/step - loss: 0.3702 - accuracy: 0.8242 - val_loss: 0.6729 - val_accuracy: 0.7388
Epoch 20/100
3/3 [==============================] - 0s 31ms/step - loss: 0.3630 - accuracy: 0.8242 - val_loss: 0.6910 - val_accuracy: 0.7400
Epoch 21/100
3/3 [==============================] - 0s 31ms/step - loss: 0.3587 - accuracy: 0.8462 - val_loss: 0.7044 - val_accuracy: 0.7267
Epoch 22/100
3/3 [==============================] - 0s 30ms/step - loss: 0.3452 - accuracy: 0.8462 - val_loss: 0.6972 - val_accuracy: 0.7279
Epoch 23/100
3/3 [==============================] - 0s 33ms/step - loss: 0.3385 - accuracy: 0.8462 - val_loss: 0.7159 - val_accuracy: 0.7255
Epoch 24/100
3/3 [==============================] - 0s 32ms/step - loss: 0.3418 - accuracy: 0.8462 - val_loss: 0.7616 - val_accuracy: 0.7267
Epoch 25/100
3/3 [==============================] - 0s 31ms/step - loss: 0.3285 - accuracy: 0.8462 - val_loss: 0.7465 - val_accuracy: 0.7304
Epoch 26/100
3/3 [==============================] - 0s 31ms/step - loss: 0.3175 - accuracy: 0.8571 - val_loss: 0.7321 - val_accuracy: 0.7654
Epoch 27/100
3/3 [==============================] - 0s 30ms/step - loss: 0.3253 - accuracy: 0.8462 - val_loss: 0.7399 - val_accuracy: 0.7557
Epoch 28/100
3/3 [==============================] - 0s 30ms/step - loss: 0.3094 - accuracy: 0.8571 - val_loss: 0.8569 - val_accuracy: 0.6989
Epoch 29/100
3/3 [==============================] - 0s 33ms/step - loss: 0.3239 - accuracy: 0.8242 - val_loss: 0.8150 - val_accuracy: 0.7255
Epoch 30/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2823 - accuracy: 0.8791 - val_loss: 0.7818 - val_accuracy: 0.7170
Epoch 31/100
3/3 [==============================] - 0s 31ms/step - loss: 0.3149 - accuracy: 0.8681 - val_loss: 0.8156 - val_accuracy: 0.7570
Epoch 32/100
3/3 [==============================] - 0s 32ms/step - loss: 0.3180 - accuracy: 0.8571 - val_loss: 0.9841 - val_accuracy: 0.6856
Epoch 33/100
3/3 [==============================] - 0s 32ms/step - loss: 0.2955 - accuracy: 0.8791 - val_loss: 0.8464 - val_accuracy: 0.7267
Epoch 34/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2828 - accuracy: 0.8791 - val_loss: 0.8204 - val_accuracy: 0.7207
Epoch 35/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2759 - accuracy: 0.8791 - val_loss: 0.9081 - val_accuracy: 0.7557
Epoch 36/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2811 - accuracy: 0.8462 - val_loss: 0.9614 - val_accuracy: 0.6917
Epoch 37/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2753 - accuracy: 0.8571 - val_loss: 0.9268 - val_accuracy: 0.7243
Epoch 38/100
3/3 [==============================] - 0s 35ms/step - loss: 0.2655 - accuracy: 0.8681 - val_loss: 0.9432 - val_accuracy: 0.6904
Epoch 39/100
3/3 [==============================] - 0s 37ms/step - loss: 0.2615 - accuracy: 0.8681 - val_loss: 0.9607 - val_accuracy: 0.6880
Epoch 40/100
3/3 [==============================] - 0s 34ms/step - loss: 0.2580 - accuracy: 0.8681 - val_loss: 0.9876 - val_accuracy: 0.6699
Epoch 41/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2573 - accuracy: 0.8681 - val_loss: 0.9870 - val_accuracy: 0.6651
Epoch 42/100
3/3 [==============================] - 0s 32ms/step - loss: 0.2562 - accuracy: 0.8681 - val_loss: 1.0309 - val_accuracy: 0.7219
Epoch 43/100
3/3 [==============================] - 0s 32ms/step - loss: 0.2536 - accuracy: 0.8571 - val_loss: 1.0502 - val_accuracy: 0.6711
Epoch 44/100
3/3 [==============================] - 0s 32ms/step - loss: 0.2489 - accuracy: 0.8571 - val_loss: 1.0319 - val_accuracy: 0.7582
Epoch 45/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2464 - accuracy: 0.8791 - val_loss: 1.0343 - val_accuracy: 0.7582
Epoch 46/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2524 - accuracy: 0.8681 - val_loss: 1.0891 - val_accuracy: 0.6832
Epoch 47/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2434 - accuracy: 0.8791 - val_loss: 1.0878 - val_accuracy: 0.6977
Epoch 48/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2445 - accuracy: 0.8681 - val_loss: 1.0765 - val_accuracy: 0.7013
Epoch 49/100
3/3 [==============================] - 0s 33ms/step - loss: 0.2439 - accuracy: 0.8791 - val_loss: 1.1374 - val_accuracy: 0.7146
Epoch 50/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2438 - accuracy: 0.8681 - val_loss: 1.1641 - val_accuracy: 0.7146
Epoch 51/100
3/3 [==============================] - 0s 32ms/step - loss: 0.2665 - accuracy: 0.8462 - val_loss: 1.2124 - val_accuracy: 0.6239
Epoch 52/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2300 - accuracy: 0.9011 - val_loss: 1.1176 - val_accuracy: 0.6929
Epoch 53/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2661 - accuracy: 0.8791 - val_loss: 1.0890 - val_accuracy: 0.7243
Epoch 54/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2432 - accuracy: 0.8791 - val_loss: 1.1934 - val_accuracy: 0.7158
Epoch 55/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2575 - accuracy: 0.8681 - val_loss: 1.2008 - val_accuracy: 0.7146
Epoch 56/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2400 - accuracy: 0.8791 - val_loss: 1.1448 - val_accuracy: 0.6904
Epoch 57/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2438 - accuracy: 0.8791 - val_loss: 1.1519 - val_accuracy: 0.6941
Epoch 58/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2411 - accuracy: 0.8791 - val_loss: 1.1633 - val_accuracy: 0.6820
Epoch 59/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2564 - accuracy: 0.8571 - val_loss: 1.2091 - val_accuracy: 0.6832
Epoch 60/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2469 - accuracy: 0.8791 - val_loss: 1.1638 - val_accuracy: 0.7243
Epoch 61/100
3/3 [==============================] - 0s 27ms/step - loss: 0.2327 - accuracy: 0.8791 - val_loss: 1.1629 - val_accuracy: 0.6880
Epoch 62/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2545 - accuracy: 0.8791 - val_loss: 1.1968 - val_accuracy: 0.6892
Epoch 63/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2380 - accuracy: 0.8791 - val_loss: 1.2883 - val_accuracy: 0.6191
Epoch 64/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2502 - accuracy: 0.8571 - val_loss: 1.3090 - val_accuracy: 0.7158
Epoch 65/100
3/3 [==============================] - 0s 27ms/step - loss: 0.2445 - accuracy: 0.8681 - val_loss: 1.2319 - val_accuracy: 0.6904
Epoch 66/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2370 - accuracy: 0.8791 - val_loss: 1.2348 - val_accuracy: 0.6965
Epoch 67/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2420 - accuracy: 0.8571 - val_loss: 1.2604 - val_accuracy: 0.7243
Epoch 68/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2358 - accuracy: 0.8681 - val_loss: 1.2566 - val_accuracy: 0.6868
Epoch 69/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2322 - accuracy: 0.8681 - val_loss: 1.2711 - val_accuracy: 0.6796
Epoch 70/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2395 - accuracy: 0.8681 - val_loss: 1.2848 - val_accuracy: 0.6771
Epoch 71/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2339 - accuracy: 0.8901 - val_loss: 1.2540 - val_accuracy: 0.6965
Epoch 72/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2379 - accuracy: 0.8791 - val_loss: 1.2360 - val_accuracy: 0.6941
Epoch 73/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2358 - accuracy: 0.8791 - val_loss: 1.3021 - val_accuracy: 0.7207
Epoch 74/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2301 - accuracy: 0.8791 - val_loss: 1.3573 - val_accuracy: 0.6856
Epoch 75/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2408 - accuracy: 0.8352 - val_loss: 1.3279 - val_accuracy: 0.6614
Epoch 76/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2308 - accuracy: 0.8681 - val_loss: 1.2865 - val_accuracy: 0.7025
Epoch 77/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2380 - accuracy: 0.8571 - val_loss: 1.2855 - val_accuracy: 0.7545
Epoch 78/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2338 - accuracy: 0.8791 - val_loss: 1.3088 - val_accuracy: 0.7473
Epoch 79/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2355 - accuracy: 0.8681 - val_loss: 1.3058 - val_accuracy: 0.7473
Epoch 80/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2356 - accuracy: 0.8681 - val_loss: 1.3164 - val_accuracy: 0.6868
Epoch 81/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2323 - accuracy: 0.8681 - val_loss: 1.3451 - val_accuracy: 0.6614
Epoch 82/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2329 - accuracy: 0.8791 - val_loss: 1.3681 - val_accuracy: 0.6941
Epoch 83/100
3/3 [==============================] - 0s 27ms/step - loss: 0.2328 - accuracy: 0.8791 - val_loss: 1.3516 - val_accuracy: 0.7449
Epoch 84/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2444 - accuracy: 0.8681 - val_loss: 1.3484 - val_accuracy: 0.7557
Epoch 85/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2420 - accuracy: 0.8352 - val_loss: 1.4151 - val_accuracy: 0.6324
Epoch 86/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2394 - accuracy: 0.8681 - val_loss: 1.3909 - val_accuracy: 0.6578
Epoch 87/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2385 - accuracy: 0.8571 - val_loss: 1.3621 - val_accuracy: 0.7025
Epoch 88/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2370 - accuracy: 0.8791 - val_loss: 1.3289 - val_accuracy: 0.7545
Epoch 89/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2400 - accuracy: 0.8791 - val_loss: 1.3342 - val_accuracy: 0.7545
Epoch 90/100
3/3 [==============================] - 0s 27ms/step - loss: 0.2345 - accuracy: 0.8901 - val_loss: 1.3759 - val_accuracy: 0.7219
Epoch 91/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2283 - accuracy: 0.8791 - val_loss: 1.3905 - val_accuracy: 0.6687
Epoch 92/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2716 - accuracy: 0.8681 - val_loss: 1.3871 - val_accuracy: 0.6626
Epoch 93/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2438 - accuracy: 0.8571 - val_loss: 1.4633 - val_accuracy: 0.6771
Epoch 94/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2430 - accuracy: 0.8681 - val_loss: 1.4174 - val_accuracy: 0.7134
Epoch 95/100
3/3 [==============================] - 0s 31ms/step - loss: 0.2454 - accuracy: 0.8571 - val_loss: 1.3184 - val_accuracy: 0.7062
Epoch 96/100
3/3 [==============================] - 0s 30ms/step - loss: 0.2354 - accuracy: 0.8791 - val_loss: 1.3106 - val_accuracy: 0.7025
Epoch 97/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2347 - accuracy: 0.8791 - val_loss: 1.3137 - val_accuracy: 0.7545
Epoch 98/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2316 - accuracy: 0.8791 - val_loss: 1.3610 - val_accuracy: 0.6784
Epoch 99/100
3/3 [==============================] - 0s 29ms/step - loss: 0.2362 - accuracy: 0.8681 - val_loss: 1.3576 - val_accuracy: 0.6252
Epoch 100/100
3/3 [==============================] - 0s 28ms/step - loss: 0.2316 - accuracy: 0.8681 - val_loss: 1.3308 - val_accuracy: 0.6953

Evaluating the Performance of Model

loss, accuracy = model.evaluate(X_test, Y_test)
print("Test Loss: ",loss)
print("Test Accuracy: ",accuracy)

26/26 [==============================] - 0s 1ms/step - loss: 1.3308 - accuracy: 0.6953
Test Loss:  1.3308433294296265
Test Accuracy:  0.6952841877937317

Prediction of Test Dataset

predictions = model.predict(X_test)

26/26 [==============================] - 0s 1ms/step

loss, accuracy = model.evaluate(X_train, Y_train)
print("Test Loss: ",loss)
print("Test Accuracy: ",accuracy)

3/3 [==============================] - 0s 3ms/step - loss: 0.2283 - accuracy: 0.8791
Test Loss:  0.22826984524726868
Test Accuracy:  0.8791208863258362

Machine Learning

X_train, X_test, Y_train, Y_test = train_test_split(x, y, test_size = 0.9 , random_state = 42)

from sklearn.preprocessing import StandardScaler
scalar = StandardScaler()
scalar.fit(X_train)

StandardScaler()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

heart_data.describe()

	Age	Sex	ChestPainType	RestingBP	Cholesterol	FastingBS	RestingECG	MaxHR	ExerciseAngina	Oldpeak	HeartDisease
count	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000	918.000000
mean	53.510893	0.789760	0.781046	0.824619	198.799564	0.233115	0.603486	136.809368	0.404139	0.887364	0.553377
std	9.432617	0.407701	0.956519	0.380500	109.384145	0.423046	0.805968	25.460334	0.490992	1.066570	0.497414
min	28.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	60.000000	0.000000	-2.600000	0.000000
25%	47.000000	1.000000	0.000000	1.000000	173.250000	0.000000	0.000000	120.000000	0.000000	0.000000	0.000000
50%	54.000000	1.000000	0.000000	1.000000	223.000000	0.000000	0.000000	138.000000	0.000000	0.600000	1.000000
75%	60.000000	1.000000	2.000000	1.000000	267.000000	0.000000	1.000000	156.000000	1.000000	1.500000	1.000000
max	77.000000	1.000000	3.000000	1.000000	603.000000	1.000000	2.000000	202.000000	1.000000	6.200000	1.000000

X_train_scaled =scalar.transform(X_train)
X_test_scaled = scalar.transform(X_test)

Logistic Regression

from sklearn.linear_model import LogisticRegression
log_reg = LogisticRegression()
log_reg.fit(X_train_scaled, Y_train)

LogisticRegression()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

log_reg_pred=log_reg.predict(X_test)
from sklearn.metrics import accuracy_score
log_reg_acc = accuracy_score(Y_test, log_reg_pred)
log_reg_acc

C:\Users\AFC\AppData\Local\Programs\Python\Python311\Lib\site-packages\sklearn\base.py:432: UserWarning: X has feature names, but LogisticRegression was fitted without feature names
  warnings.warn(

0.7762998790810157

Decision Tree Classifier

from sklearn.tree import DecisionTreeClassifier
dec_tree = DecisionTreeClassifier()
dec_tree.fit(X_train,Y_train)

DecisionTreeClassifier()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

dec_tree_pre = dec_tree.predict(X_test)
dec_tree_acc= accuracy_score(Y_test, dec_tree_pre)
dec_tree_acc

0.6807738814993954

Random Forest

from sklearn.ensemble import RandomForestClassifier
ran_cla = RandomForestClassifier()
ran_cla.fit(X_train,Y_train)

RandomForestClassifier()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

ran_cla_pre = ran_cla.predict(X_test)
ran_cla_acc = accuracy_score(Y_test, ran_cla_pre)
ran_cla_acc

0.7496977025392987

XGBoost

from xgboost import XGBClassifier
xgb = XGBClassifier()
xgb.fit(X_train,Y_train)

XGBClassifier(base_score=None, booster=None, callbacks=None,
              colsample_bylevel=None, colsample_bynode=None,
              colsample_bytree=None, early_stopping_rounds=None,
              enable_categorical=False, eval_metric=None, feature_types=None,
              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,
              interaction_constraints=None, learning_rate=None, max_bin=None,
              max_cat_threshold=None, max_cat_to_onehot=None,
              max_delta_step=None, max_depth=None, max_leaves=None,
              min_child_weight=None, missing=nan, monotone_constraints=None,
              n_estimators=100, n_jobs=None, num_parallel_tree=None,
              predictor=None, random_state=None, ...)

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

xgb_pre = xgb.predict(X_test)
xgb_acc = accuracy_score(Y_test, xgb_pre)
xgb_acc

0.7230955259975816

def print_accuracy_scores(accuracy, log_reg_acc, dec_tree_acc, ran_cla_acc, xgb_acc):
    print("Neural Network Accuracy: {:.2f}%".format(accuracy * 100))
    print("Logistic Regression Accuracy: {:.2f}%".format(log_reg_acc * 100))
    print("Decision Tree Classifier Accuracy: {:.2f}%".format(dec_tree_acc * 100))
    print("Random Forest Accuracy: {:.2f}%".format(ran_cla_acc * 100))
    print("XGBoost Accuracy: {:.2f}%".format(xgb_acc * 100))

print_accuracy_scores(accuracy, log_reg_acc, dec_tree_acc, ran_cla_acc, xgb_acc)

Neural Network Accuracy: 87.91%
Logistic Regression Accuracy: 77.63%
Decision Tree Classifier Accuracy: 68.08%
Random Forest Accuracy: 74.97%
XGBoost Accuracy: 72.31%

Neural Networks Provided the Best Accuracy Score